Für neue Kunden:
Für bereits registrierte Kunden:
Section 1: What you don’t understand can’t hurt you
Section 2: A lens to look through: OCA Theory
The US as a Benchmark
Section 3: The anatomy of FroGs: Germany as standard of comparison
Section 4: The EMUs vs. the FRoGs: New empirical evidence
Costs of a Common Currency
Section 5: Hope for the Future
FRoGs and EMUs: A look at the prospects for the European Monetary Union based on Optimal Currency Area Theory, numbers and the German economy.
Aside from a publicity campaign and a couple of large modified New Year’s Eve parties, the Euro was born on January 1, 1999 to little fanfare. Without any actual bills to enforce its reality, the general public of the 11 European nations that had just officially joined the European Monetary Union (EMU) hardly knew that anything had changed. More than a year later, as this paper is being written, one of the greatest fears of the EMU authorities is the public ignorance about the existence of the euro.
This picture is in sharp contrast to the dire warnings from economists to the negative effects of the euro on the European economy. Many argued that Europe was not homogeneous enough to support a currency union. Despite the economic benefits of lower transaction costs, decreased exchange rate variability, and the steps towards political stability in Europe, many believed the costs associated with decreased policy autonomy would make the EMU unreasonable.
While there is not enough data available to draw conclusions about whether these concerns are correct, there is reason to examine the premise of these concerns. Most of these studies are based on the Optimal Currency Area (OCA) theory, developed almost 40 years ago by Robert Mundell. They test this theory against empirical data from the prospective EMU countries and often compare it to similar results from cross-regional comparisons from the US. While these studies have a firm foundation in the OCA theory, their comparisons against American regions are somewhat inappropriate.
In this paper, I use the OCA framework to empirically test compatibility of EMU countries to the a single currency, and compare these results not to regions in the United States, but instead to German regions, which present a more appropriate standard of comparison. In section 2, I discuss the OCA theory and previous empirical studies on the EMU. In section 3, I outline how the Federal Republic of Germany (FRG) provides a better yardstick, against which to judge the viability of the EMU. Section 4 explains the empirical model I use for comparison outlines my data sources and presents the results of my regression and comparisons. Finally, in section 5, I summarize my results and add some comments as to other aspects of the study that I have ignored.
The most concise way to analyze the economic appropriateness of the EMU is through Optimal Currency Area (OCA) theory. This theory has been developing for the last 40 years and its economic foundation is very solid. The main drawback of using this theory is that it only provides a way at looking at the economic aspects of possible currency areas, not a definitive way to measure the costs or benefits. While this theory gives us an idea of economics behind the effects of a common currency, it does not tell us whether a region will necessarily be economic successful under a currency union. Still, the theory does give us a starting point from which we can compare different currency regions economically.
In 1961, Robert Mundell published the seminal article that started the discussion of OCA theory in the American Economic Review. Written in response to the various discussions across the world about common currencies and fixed exchange rates, his article laid out the economics through which academics should look at possible currency areas. Throughout the 1960s, economists McKinnon and Peter B. Kenen added two more important articles to the field. While each article focused on a different aspect of OCAs, they collectively form the basis for the OCA theory.
Mundell’s argument was based on the fact that common currency areas are a fixed exchange rate system whereby the rates cannot be nominally adjusted. Under flexible exchange rates, “depreciation can take place of unemployment when external balance is in deficit, and appreciation can replace inflation when it is in surplus” (177). In other words, when one region is in a recession, instead of suffering higher unemployment because the total output decreases, their currency can become cheaper in other areas, thereby making the goods they produce cheaper abroad. The increase in foreign demand for their products will offset the lack of demand in the domestic market, and the firms will not need to reduce employment. On the other hand, if a region experiences a boom where demand is exceeding domestic production capability, normally wages rise to induce new workers to participate in the labor market to produce the extra goods to meet demand, which causes inflation. With flexible exchange rates, an appreciation of the domestic currency will make foreign-produced goods cheaper in the booming region’s market. Consumers will then satisfy their added demand by buying foreign goods and not by forcing the domestic market into inflation. Flexible exchange rates serve as a type of automatic stabilizer to the economy. With a common currency, regions lose this stabilizer and must deal with shocks through either unemployment or inflation, both of which are socially unacceptable.
On the other hand, Mundell points to the inpractibility of having too many currencies. The former theory suggests, at its extreme, that the world should have one currency for every good in every market. In this case, money would lose all of its value as a convenient medium of exchange; this system would be hardly distinguishable from barter. In less extreme cases, having many currency brings other problems: investors face risks of valuation, consumers and firms face added costs of currency exchange, the currencies themselves are open to speculation and a degree of “money illusion” is lost. These problems become greater, the more currencies there are in the world.
Since there will be some combining of regions into currency areas, Mundell seeks to lay out a theory for which of these areas will be optimal: “optimality is here defined in terms of the ability to stabilize national employment and price levels” (178). McKinnon adds that within an optimal system, there must be balanced international payments and a stable international price level along with full employment.
The arguments against fixed exchange rates and common currencies only apply when the regions in question experience heterogeneous shocks to output. If the regions experience similar shocks the appropriate exchange rate adjustments for all regions’ currencies vis-à-vis other currencies will be the same. For example if Germany and France both experience booms and shocks together, the Mark and the Franc will both want to appreciate and depreciate against the US dollar during the same periods. Even under a flexible exchange rate system, the two countries’ currencies would not change relative to one another.
Therefore, the first criterion for an OCA is homogeneity of shocks to output.
Kenen discusses how many demand shocks are actually attributable to the dominant sector of production in the region. Therefore, if a region has a very specialized economy, it experiences the shocks to that industry in which it specializes very deeply. Moreover, its shocks coincide exclusively with shocks from other regions that are specialized in the same way. On the other hand, if regions’ economies are more diversified, then not only will shocks to specific industries be less important to the entire region’s economy, but there will also be a greater chance that the shocks of different regions will be correlated. Therefore, the similarity of the structure of the economies of different regions is an important criterion to judging their appropriateness for a common currency.
Between regions in a fixed rate or common currency system, there is the possibility of adjustments of the real exchange rate between regions, if prices and wages are flexible. If a currency has more purchasing power in one region than in another, the stabilization still takes place. In the case of a recession in one region, prices and wages in that region will go down, making goods produced there cheaper. These goods can then be sold in the other region where prices are still higher. For booming regions, a rise in prices and wages will cause goods from other regions to be relatively cheaper. When the well-off region buys products from other regions, they reduce inflationary pressure by reducing demand on the region’s production. Complete price and wage flexibility allows a fixed exchange rate or common currency area work like a flexible exchange rate system; therefore, regions with more price flexibility will be better suited to being in a common currency area.
Yet, wages in most parts of the world are by no means flexible to the extent that they can automatically stabilize the economy. In this case, factor mobility is important to stabilizing regions. By factor mobility, Mundell means the freedom for capital and investment to move to the region of highest return and the ability and willingness of labor to move to the places where real wages (wages relative to cost of living) are highest. Factor mobility, like price flexibility, can take on the stabilizing role of flexible exchange rates between two regions. When a recession hits a region, production falls. Because of decreasing returns to scale, regions in recession offer the biggest possibility of return on investment. These regions should experience a growth in investment and capital, providing for perfect capital mobility. When the region is not doing well, the possibility of losing ones job increases, and/or wages fall. Therefore, a mobile labor force will begin to leave the region for more successful regions, decreasing unemployment rates. When a region is in a disproportionate boom to other regions, interest rates will rise and decreasing returns will set in, inducing capital to flow away from the region. The increased wages and/or low unemployment will convince labor from other regions to come to where the jobs are, which holds unemployment in the booming region to its natural levels, so that inflation does not rise. Factor mobility between regions can take the place of flexible exchange rates in stabilizing regions. We must, therefore, consider the factor mobility between regions when analyzing the appropriateness of currency areas between them.
The border of currency areas should, therefore, contain regions, which have similar shocks, price and wage flexibility and/or factor mobility. Theoretically, any one of these conditions could allow for stabilization in all regions after shocks have occurred, but the best areas meet all three conditions. Accordingly, the world outside of a particular currency region should exhibit the opposite characteristics: shocks should not coincide with shocks inside the area, prices and wages should be relatively less flexible and factors should not be mobile across the borders of the currency area. Such conditions are often met within individual countries.
To summarize, Mundell outlines seven criteria that the international monetary system has to meet for flexible exchange rates to work effectively and efficiently. In other words, in order for there not to be too few nor too many currencies in the world, the following conditions must be met: 1. exchange rates must be dynamically stable to speculative demands; 2. the necessary exchange rate adjustments for normal disturbances are not large enough to cause large shift between export- and import- competing industries; 3. the risks caused by flexible exchange rates are not too large for the investors to take; 4. the central banks do not abuse their monopoly; 5. the monetary discipline of the central banks is maintained by the negative image of inflation in the market; 6. there is reasonable protection for debtors and creditors through the long run flow of capital; 7. wages and prices are not tied to a price index.
McKinnon and Mundell also note that control over monetary policy, along with independent fiscal policy, is part of a country’s economic sovereignty. Since this economic sovereignty is closely related with political sovereignty, Mundell asserts, “currencies are mainly an expression of national sovereignty, so that actual currency reorganization would be feasible only if it were accompanied by profound political changes” (180). Kenen discusses in depth the need for fiscal policy autonomy to coincide with monetary authority. “It is the chief function of fiscal policy, using both sides of the budget, to offset or compensate for regional differences, whether in earned income or unemployment rates.” (47)
Based on the OCA theory developed by Mundell, McKinnon and Kenen, many economists have attempted empirical studies to determine the possibility of success of the EMU. While some studies are economically and mathematically more robust than others, most have interesting insights into how one could quantify and empirically apply OCA requirements when faced with an actual currency union. Many studies generate numbers that by themselves have little meaning; they must be compared with a standard for monetary unions. Until now, all economists have compared results from the EMU to those from the USA. From such comparisons, they conclude that the “core” of the EMU will make a suitable currency union, but the entire 11 countries currently adopting the euro are not homogeneous enough.
Barry Eichengreen, often in conjunction with Tamim Bayoumi, has written many different studies of the appropriateness of the EMU based on empirical work. Their main contribution to this field has been through their vector autoregression (VAR) model, where they analyze the correlation of shocks to supply and demand in various proposed currency areas. They use this technique in the book One Money or Many?, which looks at various possible currency areas in the world, and in the article “Shocking Aspects of European Monetary Unification,” which looks exclusively at Europe. In these writings, they extrapolate Mundell’s theory to say that the magnitude of the losses incurred from a currency union are a function of the incidence and type of shocks to the economy and of the speed of the adjustment. They understand the most important factors in terms of OCA theory to be the nature of disturbances, timing of these disturbances and the ease of response.
In their analysis, they first look at correlation between output and inflation within possible currency areas, then go on to examine the shocks through their VAR model. When looking at different currency areas across the world, they compare business cycle correlations to the correlation of the “international business cycle” between the USA, Japan and Germany. When analyzing the EMU, they use the United States as a standard of comparison. In the former analysis, they conclude that Europe has better prospects for a currency union than other parts of the world, including North America, Southeast Asia and East Asia. In the latter study, they conclude that the core of the EMU is as closely correlated as the entirety of the United States, implying that a two-speed approach monetary unification may be appropriate.
These writings by Eichengreen and Bayoumi make two important points about the EMU. In the 1994 book, they examine the true effectiveness of monetary policy in stabilizing the economy, the implication since if monetary policy is not effective, then flexible exchange rates are less important. Monetary policy could be less important because money could be neutral in the economy, because other policies (namely fiscal policy) are available to the government, though fiscal policy poses a free-riding risk, because policy makers may misuse monetary policy, or because other factors may indicate the appropriateness of the union. In their 1993 article, Bayoumi and Eichengreen discuss the limits on fiscal policy in the EMU and how that may affect the union. Essentially, these limits may prevent a free rider problem, but they also may cause the shocks in the European economy to be greater. Governments of EMU- countries may be left without any policy options to stabilize their domestic economies after a region-specific shock. These two notes are important to keep in mind during further analysis of the EMU.
Eichengreen and Bayoumi also study the volatility of bilateral exchange rates in their 1997 article “Ever Closer to Heaven?” They found that the asymmetry of shocks in Europe was diminishing, implying that tighter economic integration was leading towards monetary integration. They also found that there is a core group of countries ready for a currency union, and then two peripheral groups: those converging towards the union and those complete unsuitable for union. However, this study must be seen in light of their comments in their 1993 article “Shocking Aspects of European Monetary Unification.” Here they argue that studying the volatility of exchange rates conflates information on the symmetry of the shocks and the speed of adjustment. Therefore, while their 1997 results are interesting, they do not seem academically useful.
One of the most recent of these OCA based empirical studies is from the European Central Bank, authored by I. Angeloni and L. Dedola. They compare correlations between output and prices in the EMU to determine the homogeneity of the European economy.
Based on Mundell and Kenen, Angeloni and Dedola claim that heterogeneous economic and financial structures lead to undesired differences that the single monetary policy will have on different regions within the EMU: “different economic structures imply asymmetries in policy transmission” (1). They compare their results to correlations between Europe, the US and Japan. They interpret their results as proof that the EMU would be best made a two-step process: there is a set of countries that satisfy the OCA requirements, and a set that are still converging. Overall, however, they conclude that the unification, while possibly bumpy, will do fine.
One criticism of their study is that the numbers are overwhelming and inconclusive. They present tables of numbers that are hard to read and harder to understand. Moreover, when compared with the baseline numbers for US-Japan-Europe correlation, the numbers are less conclusive. A critical reader must wonder if authors publishing for the ECB concluded the soundness of the EMU for other than economic reasons. Still, the methodology used in the study is sound.
Antonio Fatas presented the article “EMU: Countries or regions?” in which the arguments are sound, but the study is somewhat weak. He argues that, based on OCA theory, when national business cycles are pronounced, exchange rates must be flexible to stabilize the economy, especially in Europe, where prices and wages are sticky. He studies the unemployment trends across Europe and concludes that the importance of national borders in Europe is decreasing, but the distinction between regions is increasing. He asserts that European economic integration has increased cross-border correlation, but decreased within-border correlation of the economy. His study of the unemployment rates of the EMU countries suffers from the same difficulties as Eichengreen and Bayoumi’s study of exchange rates: it analyzes not only the shocks, but simultaneously the speed of response to the shock. Without separating the type of shocks from the speed of response, Fatas cannot claim to have a reliable conclusion about the suitability of the EMU to a common currency. Still, Fatas’ assertion about the importance of business cycles is important.
Finally, Ramaswamy and Sloek look at the differences within EMU countries of the effects of monetary policy on economic activity. Their argument, similar to that for the importance of business cycles in OCA theory is that countries need to not only have similar shocks and therefore require similar monetary policy, they should also respond similarly to that monetary policy in order for the monetary union to work best. They find that the EMU countries fall into two groups according to their responses to monetary policy, and that these groups coincide, with a few notable acceptations, to the commonly held “core” and “periphery” of the EMU. The differences between the groups are important: changes in economic activity in response to monetary policy take twice as long to occur and go twice as deep in one group compared to the other.
There are many other similar empirical studies on the appropriateness of the EMU according to OCA theory, many comparing the EMU-area to the USA. These studies often conclude that the “core” euro-area countries tend to be similar to the USA as a whole in terms of OCA requirements, and therefore will make a proper currency union.
Paul Krugman uses the American experience with a common currency to draw some lessons and illustrate warnings for Europe. He asserts that factors within the US single currency system allow regional crises to happen and magnify. Greater economic integration leads to greater specialization in regions, which makes regions vulnerable to industry-specific shocks becoming regional crises. Because of this specialization, interregional exports become less stable, which also allows for asymmetric regional shocks. Furthermore, with capital mobility, investment in the US flows pro-cyclically, deepening asymmetric shocks. The effects of these factors are exhibited by the divergent long-term growth rates in US regions.
Since these problems stem from factors in the monetary system, the regional problems must be solved through federal fiscal transfers: “within a monetary union stabilization policy must take the form of fiscal intervention” (255). Regional policy, however, is ineffective in the USA, because it acts pro- cyclically. Almost all states have balanced-budget laws, which take away the automatic stabilizer of carrying a deficit during recessions. Even if the regions could act counter-cyclically, the market prevents them from it. Because of their relative small size, states in the USA face a higher cost of borrowing as they carry more debt. Since monetary and fiscal policy cannot act against a regional recession in the US, the only way for an American region to reduce unemployment in a downturn is to reduce its labor force. Krugman points out that Europe’s extraordinarily low labor mobility excludes this option from the EMU, and causes serious concern over the possibility of permanent, deep regional recessions.
Eichengreen also compares the European situation to the American in his 1997 article “One money for Europe? Lessons from the US Currency Union.” He claims that the success of the US common currency is that American regions never speak of succeeding from the Federal Reserve System. He agrees with Krugman that there are considerable regional differences within the United States, as exhibited by the conflicts between the District Banks within the Federal Reserve. These differences are mitigated by the federal transfer system, a fact that does not bode well for the EMU, where the total fiscal transfers are less than 1% of total GDP. This federal system serves to address some of the economic externalities that affect the entire country, as well as a type of insurance against economic downturns.
He does observe some harmonization within the American currency union, especially in tax rates. Although firms do not exclusively base their location decisions on tax rates (which allows tax rates to be diverse in a competitive economy), there are spillovers to fiscal policy and taxation that induce fiscal policy coordination between US states.
The lessons from the American experience with a common currency are certainly valuable for the EMU. On the one hand, they give academics a benchmark against which they can judge the euro-area’s chances for prosperity and more generally the importance of difference factors within OCA theory. On the other hand, they are full of lessons for European policymakers and economists as to what to expect from and how to respond to the EMU. At the same time, these studies point out the differences between Europe and the USA, which brings into question the value of these comparisons.
Various authors question the value of the USA as a benchmark for Europe. In their discussions to Bayoumi and Eichengreen’s “Shocking Aspects of European Monetary Unification,” Giorgio Basevi and Patrick Minford both criticize the article on this basis. Both authors understand the problems of comparing the American system to the EMU because of their economic structural differences. Especially in terms of factor mobility, and specifically labor mobility, which is a key factor in OCA theory and in the theory behind many of these empirical studies, the USA is very different from the EMU. It seems that Krugman’s, Eichengreen’s and similar studies can shed light on general OCA theory and on some policy issues for the EMU, but because of the differences between the USA and the euro-area, the studies do not use the appropriate benchmark for comparison.
A better standard of comparison would be Germany. Not only is Germany a federal republic, in which the Bundesländer have heterogeneous economic systems and some fiscal autonomy, the underlying structure of the German economy more closely resembles that of the structure of the EMU on the whole. Within this structure, the aspects of the EMU’s economy that are integral to OCA theory - factor mobility, wage and price flexibility and economic diversity - are more similar to Germany than to the USA. While there are still important differences between the German economy and that of the EMU, especially in terms of fiscal policy, the overall structure of Germany is more closely related to the euro-area economy than the American system is. Therefore, Germany is a better standard of comparison than the USA.
In terms of heterogeneity of regions, the German economy strongly resembles the EMU-area. Both areas contain larger regions, which are relatively prosperous, yet somewhat specialized in terms of output - in Germany, Bavaria and Baden- Württemburg; in Europe, France and Germany. Each area also has smaller regions to which trade is very important and migration is much higher. In Germany, these are the city-states of Hamburg, Bremen and Berlin; in the EMU, the small nations of Luxembourg and to a lesser extent Belgium, the Netherlands and Portugal. The geographic nature of the EMU and Germany and the similarity of their regional structure make Germany a good benchmark for the EMU. Within the structural aspects of the OCA theory, factor mobility is the area where German is most obviously a superior benchmark to the USA. Factor mobility is the ability of capital and labor to flow freely between regions. Since the Maastricht Treaty, capital flows within the European Union are as free as within any of the member states. Consequently, capital regulation between EMU countries must exactly resemble capital flows within German regions. While capital mobility is also very high in the United States, the fact that almost the exact same regulations apply to investment within Germany as within the euro-area implies that German’s capital flows even more closely resemble those in the EMU.
Labor mobility is the largest obstacle to European factor mobility, because of the language and cultural differences that exist on the Continent, as well as the historically low migration rates there. In the US, it is much more common for families to move away from their tradition home state to find a job or to earn more money. Studies show that American labor mobility is three times greater than labor mobility within France and Germany (Eichengreen 1997). While labor mobility within Europe as a whole is harder to measure, it is obvious that it is even lower than within the individual nations. Still, Germany stands as a better point of reference than the US in terms of labor mobility. Even though Germany does not exactly resemble the European economy, it comes closer to doing so than the United States.
In terms of price and wage flexibility, Germany again does a better job of modeling the EMU than the US does. While prices seem comparably flexible in all three areas, wages in the EMU and Germany and more likely to be based on a union agreement. Since unions are so much stronger in Europe than they are in the United States, they have much more bargaining power. Despite significant differences, the foundational structure of unions across Europe is the same, and the tactics for organizing labor are similar. In fact, with increased European integration, European unions are working together more often and sharing more information. Through this integration process, the wage structures across Europe are becoming more similar.
The situation in the United States is quite different. The unions do not have nearly as much power, either in terms of the labor market or in terms of government influence. Therefore, wage structures in the USA tend to be different than in Europe and the wages in general tend to be more flexible.
Since the underlying structure of wages is the same and becoming more similar in Germany and in the EMU as a whole, the German economy can be said to have similar wage flexibility to the euro-area. In any case, the wage and price flexibility in EMU is much more like that in Germany than in the US.
There are still important differences between the structure of the German economy and the structure of the entire euro-area. Some of these differences, like labor mobility and diversification of the economy, are likely to be reduced trough continued economic integration, while some of these differences, most notably the fiscal transfer system, remain a concern in comparing Germany to the EMU without qualification.
While wage and price flexibility are very similar in the two areas, factor mobility within Germany is certainly greater than within the entire EMU, if only because of language barriers.
A comparison of German Bundesländer also shows that they are more economically specialized than EMU countries, implying that the comparison between the euro-area and Germany is not perfect. But it is reasonable to assume that with monetary integration and the continued process of EU general economic integration, that specialization will increase within the euro-area to more closely resemble Germany, and that factor mobility, especially labor mobility will at least begin to increase and possible converge to near-German levels. Therefore, the long-term projections for the EMU may be extrapolated from a comparison with Germany.
The most important difference between the German economic system and that of the EMU is the federal transfer system. Germany’s fiscal transfers are even more equalizing than the federal system in the United States. According to the Constitution, taxes must be divided between the Bundesländer so that no Land has a tax capacity of less than 92% of the national average tax capacity. Through such a radical system, the German Republic ensures that no region will fall too far behind the others. The EMU has no such system. As mentioned before, less than 1% of euro-area GDP is spent on fiscal transfers between regions. Moreover, there is no automatic system for these transfers to be allocated to regions in recession. The bulk of these funds are funneled through specific assistance programs for the worst performing regions in the EMU. Fiscal federalism in Germany acts much faster and more radically to correct regional imbalances than the EMU system.
The difference in the fiscal transfer system is especially in terms of OCA theory, because fiscal policy can be used to stabilize regions experiencing asymmetric shocks when monetary policy is restricted through a common currency. Fiscal policy will be needed to spur economies in recession when there cannot be an exchange rate adjustment to make their products relatively cheaper. The EMU countries retain autonomy over their domestic fiscal policies1, but in light of Krugman’s argument that regional fiscal policy can only act pro-cyclically, this policy may not be enough. The fiscal transfer system, therefore, must be automatic and large enough to offset regional shocks and stabilize the entire economy.
Since the EMU’s economy is structurally similar to the German economy, the results obtained from OCA empirical studies that compare the euro-area to Germany are much more likely to paint a realistic picture of the prospects for monetary unification. If the results are similar to Germany’s in terms of the type or correlation of shocks within the EMU, then one can assume that the basic structure of the economy in both areas will generally respond similarly to the shocks and that the shocks will have similar effects on the regions. The differences in fiscal policy and labor mobility must be kept in mind during the analysis: the EMU is still not as integrated as unified Germany.
There are other caveats to this argument. Most notably is that the German economy is not performing especially well, and has not been since unification. Many ask whether a unification of former East and West Germany is actually an optimal currency These restrictions also encourage “regional” fiscal policy within the EMU to act pro-cyclically. area by itself (see Ghosh and Wolf). Yet, the argument still holds when comparing the EMU to the better performing West German economy before 1970. Furthermore, the German currency union is still an economic fact and its existence has not caused the German economy to completely collapse. The economic drawbacks of unification have not overwhelmed the political and social benefits; in public opinion, having one German currency and a unified German economy is still very popular. These facts lend hope to the euro’s performance in the short run. If the EMU does not perfectly match the OCA theory, having a common currency will not necessarily burst the members’ economies. Even when the economic costs remain somewhat high, the other benefits of the single currency will allow the monetary union to continue.
That said, the next step is to look at both Germany and the EMU through the OCA theory empirics and compare them.
As we have already seen, there are many different ways to empirically apply OCA theory to the EMU. There are three main aspects through which we can study a proposed currency area: 1.
The underlying shocks to the economy; 2. The availability of structural stabilizers to the economy and their effects on the economy; and 3. The structural similarity of the regions within the currency area.
In examining shocks, we are trying to determine how asymmetric the shocks to various regions actually are. We can look at incidence of shock, to see how many region-specific shocks there are, we can look at magnitude of shocks, to examine how important the shocks are, and / or we can look at type of shocks, to analyze both how asymmetric the shocks would be to specialized economies2 and how long the shocks usually affect the economy.
In looking at the structures that stabilize the economy after shocks have occurred, we can conclude whether the economy will be able to remain stable in a single currency system. These stabilizers include price and wage flexibility, factor (capital and labor) mobility and fiscal policy. We can measure the relative importance and effect of each stabilizer in adjusting from shocks and the speed in which the stabilizer leads to adjustment. We can also conflate the information on the type of stabilizer and analyze the general response of the economy to shocks, including the speed and magnitude of adjustment. By conflating, information, we can examine at the total cost to the economy because of the loss of monetary sovereignty and the inefficient working of stabilizers.
By analyzing the structural similarity of the regions within an area, we can extrapolate not only possible asymmetry of shocks based on the structure of the economy, but also the possibility of the stabilizers to work symmetrically in all regions. Different economic structures between regions imply that they will experience different types of shocks, and that they will respond differently to both the shock and the policy applied to remedy the shock. In a common currency area with a common monetary policy, it is important that the policy adjustments made to remedy common shocks to the economy3 affect all regions similarly. If they do not, a symmetric shock across all regions may effectively translate into an asymmetric shock, causing regional differences, tension within the currency union, and possible unemployment or inflation in some regions.
I study two aspects of the European Monetary Union that fall into the second and third categories above. First, I look at the correlation of business cycles within the euro-area, and then within Germany. Business cycle correlation implies similar economic structure and similar response to policy. Moreover, high correlation in the business cycle means that the different regions will be in recession or boom at the same time (effectively facing symmetric shocks). If so, they will desire the same monetary strategy, allowing the central bank to follow the popular logic in deciding policy. If the business cycles are less correlated, the economy will have to rely more on the other factors in OCA theory to stabilize the economy: price flexibility, factor mobility or fiscal policy.
Secondly, I look at the costs associated with joining a common currency as seen as a loss caused by the difference between the optimal exchange rate for a region and the optimal exchange rate for the currency area. Since the exchange rates for the area are determined by an optimality condition for the entire area, some states suffer in a loss in output because what is optimal for the entire area is not its optimal choice. This loss will have to be recovered either through the stabilizers taking the place of exchange rates or at the cost of the economy on the whole.
The data for the European Monetary Union was found in the International Financial Statistics database, published by the IMF. The information for real GDP was not uniform, so I used exclusively nominal GDP numbers. Regional data for Germany proved harder to find. After much searching, I received the bulk of the data for German Bundesländer between 1970 and 1990 from Prof. Wolf. The data was originally collected from the Landesamt für Statistik Baden-Württemburg, and contains both nominal and real GDP information. For regional data from 1990 to 1998 for unified Germany, I looked to the Landesamt für Statistik Saarland, who kindly published by nominal and real GDP data on their website.
The study of the business cycles is very straightforward. For each region (Bundesländer or EMU Member States) I ran a regression of the nominal GDP on a constant and the GDP from the previous year. These results gave me the general growth trend for each region. The residuals from each regression represented the part of the business cycle in which the region found itself for that year: they are the difference between the actually GDP and the projected GDP based on the trend. These values included both cyclically factor of the economy and shocks that affected output.
Finally, I calculated the correlation of these residuals for every pair of regions in the currency area. These correlations represent roughly the relationship between the cyclically aspects of the economies of the regions.
Since the residuals used for correlation also contain some shock information, the correlation coefficients also conflate information about the correlation of shocks to output between different regions. Therefore, a comparison of correlations includes not only how closely the business cycles in different regions follow each other, but also how related shocks to the economy are. This fact does not totally eliminate the validity of this type of study within OCA theory: ultimately, both the business cycles and the shocks in different regions should be perfectly correlated. It does have, however, implications for comparing different types of areas.
We should expect that these correlations are higher in Germany than in the EMU. The Federal Republic of Germany has been unified for the last ten years and West Germany for the past 130. Not only should the economy be more similar because of its history of unification, but also this similarity should be amplified by the conscious, rigorous agenda of economic integration that the German government has pursued in the past ten years. Finally, the federal fiscal transfer system should also cause the correlations to be higher, because it automatically transfers income from the regions in a boom to those in a recession. In doing so, it correlates and controls the business cycles in the regions.
My results show that the correlations between business cycles in the EMU are, in fact, significantly lower than those in the German Bundesländer. The EMU correlations are presented in the Appendix in Table 1. The negative correlations are printed in red and all correlations greater than 0.50 are highlighted with a gray box. Interestingly, the average of all unique correlations is 0.32. Excluding the negative values, the average is 0.42. Thirty six percent of the correlation pairs have a value greater than one half.
Though these statistics seem to indicate that the euro-area economies are at least partially correlated, they do not present a clear picture of the possibility of success for the EMU. To draw any conclusions about their significance, we need to compare these numbers with the results of the analysis of the Federal Republic of Germany.
The current FRG seemed to be less regionally correlated than the EMU area. The entire set of pair-wise correlations for Germany is presented in Table 2 in the Appendix. Interestingly, the average of these correlations is only 0.20. Excluding the negative correlations, the mean is much higher, 0.62. The negative correlations are also significant in value; their average value is -0.40. Out of the 120 unique correlation pairs for the Bundesländer, 39% have values greater than 0.50. The German regions seem to be either highly positively or negatively correlated with each other, but rarely weakly correlated.
This division in correlation could be due to the recent reunification of East and West Germany. This economic shock affected the two sides differently. The correlations between the current FRG, former West Germany and former East Germany are presented in Table 3 below.
Abbildung in dieser Leseprobe nicht enthalten
The West is strongly correlated with the macro-business cycle of the current FRG, while the East is strongly negatively correlated with both the West and the current total. A closer examination of Table 2 reveals that the former Western Länder are positively correlated with each other, as are the former Eastern regions. Looking exclusively at former West Germany, as illustrated in Table 4 in the Appendix, there are no negatively correlated Bundesländer. The average correlation is also much higher in the West: 0.72. Almost all of the regions are highly correlated with each other; 95% of the bilateral correlations have a value greater than 0.50.
Correlations for the former Eastern regions, shown in Table 5 in the Appendix, show that they also have no negative correlations. Their residuals are more weakly correlated than those in West Germany: 0.61. In addition, only 73% of the bilateral correlations in the East have values greater than 0.50.
The basic correlation statistics for the four examined areas - the EMU, the current FRG, former West Germany and former East - are presented in Table 6 below. The mean correlation over the entire EMU is higher than that over the entirety of the current FRG. In other words, on the whole, the business cycle and to a lesser extent the shocks to output are more correlated currently in the EMU than in the FRG. The average of the positive correlations is higher in the FRG than in the EMU; of the regions that have similar business cycles, the actions of the cycle is more correlated in Germany than in the euro-area. Both the mean of all correlations and that of only the positive ones is less in the EMU than in either former West or East Germany.
Table 6: Basic Correlation
Abbildung in dieser Leseprobe nicht enthalten
The standard deviations presented in the table represent how dispersed the correlations between regions actually are. In economic terms, they indicate how different the relationships between business cycles of different regions actually are. The euro-area has a similar dispersion of correlation to former East Germany, which is lower than that for the current FRG but higher than for former West Germany.
Finally, the percentage of bilateral correlations with values greater than 0.50 is roughly the same in the EMU and in the FRG. In both areas, it is smaller than in both former West and East Germany. This fact indicates that there are more pairs of regions with strong relationships between their business cycles in the former divided Germany. The EMU again looks similar to the current FRG in this respect.
Since the asymmetry of present-day Germany is certainly due, in part, to unification, this implies that without this onetype large shock, Germany would be more correlated than Europe on the whole. Yet, Europe has begun a rigorous integration program that parallels Germany’s. There is definitely hope that the euro-area correlations can grow over time, making the euro-zone a more appropriate currency area.
By this measure, the EMU seems to be as appropriate a currency area as unified Germany in terms of the business cycle correlation and symmetry of shocks between regions. We must take into account that Germany still has economic means to deal stabilize asymmetric shocks that are unavailable to the EMU. Labor mobility is still somewhat higher and federal fiscal transfers make up an important part of the economy. These means of stabilization will likely become more available to the euro- area with increased integration. Overall, the correlation of the business cycle across euro-area regions seems to indicate that the EMU has a good chance at success.
Another way to examine the suitableness of an area for a common currency is by calculating the costs associated with adopting the currency. Artish Ghosh and Holger Wolf develop a method for calculating these costs in their paper “On the mark(s): Optimum currency areas in Germany.”4 They argue that monetary policymakers in common currency areas adjust policy to offset the composite shock to the economy. When regions experience shocks that require greater adjustment than monetary policy officials are willing to make, they suffer costs that must be offset through means other than monetary policy.
In terms of OCA theory, this argument says that the differences between shocks to a particular region and those to the entire area will translate into real costs for the regions.
There will also be benefits from having a common currency, but these must outweigh the costs to make the currency union viable. Moreover, since the other methods with which the economy could stabilize itself are limited in Europe, these costs become more serious. Specifically, Ghosh and Wolf mention price and wage rigidity and low labor mobility as factors that increase these costs within Europe.
Ghosh and Wolf develop an empirical model to test the appropriateness of different areas to a single currency. They first calculate the general adjustment that each region would need to complete offset economic shocks. These shocks are represented by innovations in the growth process calculated in the error term from the regression of output in the current period for each state on a constant and the output from the previous period. They then calculate the average adjustment for the area by weighting each particular regional adjustment by regional output. The difference between regional optimal adjustment and the area-wide average adjustment will be the cost of the adopting the currency area for the region.5 The total cost for the region is calculated by weighting these regional costs by regional output.
Using the data described above, I performed these regressions and obtained costs for the EMU and the current FRG. As a standard of comparison, I calculated costs for former West Germany between 1970 and 1990 and between 1970 and 1998 and for former East Germany between 1990 and 1998. The latter empirical works updates Ghosh and Wolf’s work to include data from unified Germany. Table 7 below presents the results for average cost for the entire region for the five regions examined.
Abbildung in dieser Leseprobe nicht enthalten
The first observation is that the EMU and the current FRG seem to suffer the same costs from a currency union during the periods analyzed. These costs are less than those experienced by East Germany since unification with West Germany, but greater than those suffered by West Germany. The costs to West German regions before unification were less than half of those experienced by the current FRG and the EMU.
These results imply that the euro-area will experience similar costs in monetary union to those costs currently experienced in Germany. While these costs are still higher than the benefits estimations by Ghosh and Wolf and published by the European Union of savings of 0.5% of GDP across the euro-area, they are optimistic. Obviously, there are other factors determining the currency union in Europe than pure economic costbenefit analyses. While extraordinary costs could outweigh those other considerations, there is no reason to believe this will happen in Europe. The costs seem to be in line with those experienced by Germany during its unification. Again, comparing the EMU with Germany gives a positive prognosis for its survival. Though both areas suffer costs of currency union, Germany has been able to carry them without major problems for the last 10 years. There is no reason to believe the EMU will not be able to carry similar costs.
This paper examines the European Monetary Union in terms of Optimum Currency Area theory and compares the results to those obtained from examining Germany. OCA theory provides the most developed, concise and economically sound way to analyze the proposed currency union in Europe. Germany provides a better standard of comparison for Europe than the USA because it closer resembles the euro-area, especially in terms of the OCA factors of labor mobility and wage flexibility. Germany is not the perfect benchmark of comparison, especially since its economy has been integrating for much longer than Europe’s and it has a developed system of federal fiscal transfers. Still, it is better than the USA and does provide a starting point.
My empirical analysis showed that the structural similarity, the symmetry of shocks and the possible costs of a currency union are quite similar in the EMU as they are in the current FRG. Correlations of business cycles across EMU member states looked surprisingly similar to those across German Bundesländer, in both magnitude and dispersion. Moreover, the costs of entering the currency union are potentially the same for the euro-area as for the current FRG. The German system might be able to deal with asymmetry and costs better than the EMU, because of their slightly higher labor mobility and because of their federal transfer system. Since the further integration of Europe promises to bring both of these means of stabilization to the EMU, the prospects still look good in the long run.
Using Germany as a benchmark to judge the EMU is not only more theoretically sound than using the USA, it provides more optimistic results. While my studies were somewhat limited in this paper, there are still plenty of models to be applied to this area. Furthermore, as Germany has recently had experience with rapid economic integration, there are many lessons from the FRG that can be applied to the EMU. When can only hope that positive lessons and attitudes can continue to allow the euro’s path to be smooth.
Table 1: Correlations of Business Cycles in the EMU
Abbildung in dieser Leseprobe nicht enthalten
Table 2: Correlations between German Bundesländer
Abbildung in dieser Leseprobe nicht enthalten
Table 4: Correlations in former West Germany
Abbildung in dieser Leseprobe nicht enthalten
Table 5: Correlations in Former East
Abbildung in dieser Leseprobe nicht enthalten
Angeloni and Dedola 1999: “From the ERM to the Euro: New Evidence on Economic Policy and Convergence Among EU Countries,” ECB Working Paper #4.
Basevi, Giorgio 1993: “Discussion” to Eichengreen in Torres and Giavazzi (eds) Adjustment and Growth in the European Monetary Union. Cambridge University Press.
Bayoumi Tamim and Barry Eichengreen 1993: “Shocking Aspects of European Monetary Unification,” in Torres and Giavazzi (eds) Adjustment and Growth in the European Monetary Union. Cambridge University Press.
Bayoumi, Tamim and Barry Eichengreen 1994: One Money or Many? Analyzing the Prospects for monetary Unification in Various Parts of the World. Princeton Studies in International Finance No. 76 September 1994 (Princeton, NJ: Princeton University).
Bayoumi, Tamim and Barry Eichengreen 1996: “Operationalizing the Theory of Optimum Currency Areas”, CEPR discussion paper #1484.
Bayoumi, Tamim and Barry Eichengreen 1997: “Ever Closer to Heaven? An optimum-currency- area index for European countries” in European Economic Review 41 761-770.
Blanchard, Olivier and Danny Quah 1989: “The Dynamic Effects of Aggregate Demand and Supply Disturbances,” American Economic Review, 79 (September 1989), pp. 655-673.
Blanchard, Olivier and Lawrence Katz 1992: “Regional Evolutions,” Brookings Papers on Economic Activity, No. 2 (1), pp. 1-61.
De Grauwe, Paul and Wim Vanhaverbeke 1993: “Is Europe and Optimal Currency Area? Evidence from Regional Data,” in Paul Masson and Mark Taylor, eds. Policy Issues in the Operation of Currency Unions. (Cambridge and New York: Cambridge University Press) pp. 111-129.
Eichengreen, Barry 1997a: “One money for Europe? Lessons from the US Currency Union,” in Eichengreen, European Monetary Unification. (Cambridge, Massachusetts: MIT Press) pp 13-50.
Eichengreen, Barry 1992: “The Effects of the ERM on Participating Economies,” International Monetary Fund Staff Papers, 39 (June 1992), pp. 330-356.
Eichengreen, Barry 1997b: “Is Europe an Optimal Currency Area” in ibid. pp. 51-72.
Eichengreen, Barry 1997c: “Labor Markets and European Monetary Unification, “ in ibid. pp. 121-152.
Fatas Antonio 1997: “EMU: Countries or regions? Lessons from the EMS experience” in European Economic Review 41, 743-751.
Ghosh, Artish R., and Holger C. Wolf, “On the mark(s): Optimum currency areas in Germany.” Economic Modelling. 13 (1996) pp. 561-573.
Hartmann, Jürgen (ed.) 1997: Handbuch der deutschen Bundesländer. Handbook of the German States. 3rd Edition (Frankfurt, Campus Verlag).
Kenen, Peter B. 1969: “The Theory of Optimum Currency Areas: An Eclectic View”, in Robert
A. Mundell and Alexander K. Swoboda (Eds) Monetary Problems of the International Economy. Chicago University of Chicago Press, pp 41-60.
Krugman, Paul 1993: “Lessons of Massachusetts for EMU,” in Torres and Giavazzi (eds) Adjustment and Growth in the European Monetary Union. Cambridge University Press.
Laufer, Heinz and Ursula Münch 1998: Das föderative System der Bundesrepublik Deutschland. (The Federal System of the Federal Republic of Germany) (Augsburg, Leske+Budrich, Opladen).
Minford, Patrick 1993: “Discussion” to Eichengreen in Torres and Giavazzi (eds) Adjustment and Growth in the European Monetary Union. Cambridge University Press.
Mundell, Robert A. 1961: “A Theory of Optimum Currency Areas”, American Economic Review n. 51 pp 657-665. Adapted in International Economics. (New York: Macmillan, 1968) pp. 177-186.
Organization for Economic Co-operation and Development 1987: Flexibility in the Labor Market. Paris, OECD.
Poloz, Stephen, S. 1990: “Real Exchange Rate Adjustment Between Regions in a Common Currency Area,” Ottawa, Bank of Canada, processed.
Ramaswamy and Sloek 1997: “The Real Effects of Monetary Policy in the European Union: What are the Differences?”, IMF Working Paper (pp 1-12.)
Tavlas, George S. 1992: “The ‘New’ Theory of Optimal Currency Areas,” Washington, DC IMF, processed.
Villaverde, Jose and Blanca Sanchez-Robles 1999: “The European Union and Regional Convergence in Spain: A New Approach.” In Meeusen, Wim (ed) Economic Policy in the European Union: Current Perspectives. (MA, USA: Edward Elgar Publishing, Inc).
Welfens, Paul J. J. (ed) 1997: Economic Aspects of German Unification:Expectations, Tranistion Dynamics and International Perspectives. 2nd Edition. (Berlin: Springer-Verlag).
1 Expect for the Stability and Growth Pact, which limits EMU member states to a deficit no larger than 3% of GDP and a debt no larger than 60% and encourages tighter fiscal coordination.
2 This argument especially makes sense in light of the fact that increased monetary and economic integration almost always leads to increased specialization. Looking at types of shocks may be especially appropriate for determining whether an area that is currently under a flexible exchange rate regime and not integrated will be suitable for a common currency if integration takes place.
3 Or any other common shock that affects the country as a whole.
4 They look specifically at whether unified Germany comprises an OCA and conclude that the costs are too high to warrant the benefits. From their analysis, not even former West Germany alone was suitable for a common currency because of the asymmetry of shocks and because of the diversity of the economy. The conclude that the currency union and integration of the two German economies could not have been a purely economic decision.
5 This only holds when the difference is greater than zero; only if the region requires greater monetary adjustment than the area will make, will the region suffer the loss. Otherwise, the cost is zero.